The innate immune response is clearly an integral part of the secondary injury cascade in CNS trauma, but the role of glial and monocyte derived pro-inflammatory cytokines is complicated by multiple downstream effects mediated by multiple concentration-dependent receptors, and a rapidly changing and evolving microenvironment. We have new findings that strongly support the hypothesis that TNF and AMPAR changes are critical in both secondary injury and recovery after CNS injury. Sequestering TNF using soluble TNF receptor protein (sTNFR1) reduces damage after cervical SCI, and, in a highly dose-dependent manner, improves neurological outcomes. We propose to extend our studies of TNF and AMPAR trafficking as a therapeutic target for SCI using a multivariate approach to test preclinical efficacy. Etanercept (ETAN) is a TNF-sequestering protein biologic used clinically in rheumatoid arthritis; topiramate (TPM) is a neuroprotective, anti-epileptic drug that has AMPAR antagonism. We will use these drugs to target AMPAR-trafficking and AMPAR activity, separately and in combination. We will systematically evaluate the biological responses to these drugs using early post-injury biomarkers that have predicted neurological outcomes in our prior work. We will use the biomarker data to plan preclinical dose and timing regimens to evaluate effectiveness in both cervical and thoracic rat SCI, and evaluate their effects on autonomic, sensory and motor outcomes. These efforts are aimed at moving towards clinical application of anti-TNF therapies for SCI, and may be applicable to other CNS degenerative disorders as well. We propose three specific aims: Aim 1: We will evaluate the effects of ETAN and TPM on the time course and extent of biomarkers of AMPAR surface expression, cell death and the production of pro-inflammatory cytokines after cervical SCI. We predict that these treatments will reduce the feed-forward cascade of cell death. Aim 2. Guided by these data, we will optimize dose and timing of single and combination therapies to maximize six week recovery after unilateral cervical SCI using a variety of forelimb functional tests (grooming, paw placement, Catwalk and IBB). Aim 3: We will test the effects of optimized drug regimens on recovery from thoracic contusion lesions using a battery of tests that includes autonomic, sensory, and motor outcomes. This will establish whether efficacy extends to multiple models of SCI. TNF and AMPARs are also involved in the production of chronic hypersensitivity after nerve injury (Choi et al, 2010), and we will test whether this occurs after SCI as well. TPM is already used to treat chronic SCI pain. We predict that acute treatments with ETAN and TPM that result in better motor outcomes will also result in reductions in long term allodynia and in tonic, aversive central pain (King et al, 2009), the latter measured by place preference tests.